uucore: for fs use real inode dev stat on wasi

[eduardomourar]

FileInformation previously fell back to a std::fs::Metadata-based heuristic (comparing file type and size) on WASI, since it lacked access to nix::sys::stat. Switch it to rustix::fs::Stat, the same backing type used on Unix, giving WASI real inode/device identity, nlink, and equality checks instead of an approximation. Also add WASI implementations of are_hardlinks_to_same_file and are_hardlinks_or_one_way_symlink_to_same_file via rustix::fs::stat/ lstat, since std::os::unix::fs::MetadataExt is unavailable there.

[codspeed]

Merging this PR will regress 2 benchmarks

⚡ 17 improved benchmarks
❌ 2 regressed benchmarks
✅ 380 untouched benchmarks
⏩ 54 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	sort_dictionary_order[500000]	1.9 s	2.1 s	-6.54%
❌	Simulation	rm_recursive_tree	13.7 ms	14.4 ms	-5.02%
⚡	Simulation	three_39_bit_primes	545.9 ms	405.7 ms	+34.56%
⚡	Simulation	cut_fields_custom_delim	65.9 ms	51.9 ms	+27%
⚡	Simulation	cut_fields_tab	57.4 ms	45.5 ms	+26.24%
⚡	Simulation	cut_bytes	18.1 ms	15.5 ms	+16.69%
⚡	Simulation	check_sorted_utf8_locale	491.1 ms	422.9 ms	+16.13%
⚡	Simulation	five_38_bit_primes	1.7 s	1.5 s	+15.04%
⚡	Simulation	sort_numeric_utf8_locale	40.8 ms	36.1 ms	+12.99%
⚡	Simulation	merge_single_file_utf8_locale	133.4 ms	119.3 ms	+11.84%
⚡	Simulation	cut_characters	25.4 ms	22.9 ms	+10.88%
⚡	Simulation	merge_pre_sorted_files_utf8_locale	251.9 ms	238.1 ms	+5.77%
⚡	Simulation	merge_pre_sorted_files	252.5 ms	238.8 ms	+5.72%
⚡	Simulation	sort_german_de_locale	295 ms	282.1 ms	+4.6%
⚡	Simulation	sort_case_sensitive[500000]	334.9 ms	321 ms	+4.32%
⚡	Simulation	sort_spill_to_tmp_files_utf8_locale	332.1 ms	321 ms	+3.44%
⚡	Simulation	sort_reverse_utf8_locale	81.9 ms	79.2 ms	+3.43%
⚡	Simulation	sort_mixed_utf8_locale	82.1 ms	79.4 ms	+3.4%
⚡	Simulation	sort_unique_utf8_locale	84.9 ms	82.2 ms	+3.34%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing eduardomourar:wasi-fs-inode-dev-stat (be1501f) with main (8f6a838)}

Footnotes

1. 54 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/date/resolution (fails in this run but passes in the 'main' branch)

[eduardomourar]

@sylvestre, this one is ready (now with proper testing in place).

[sylvestre]

Thanks, this looks much better: there's one rustix impl shared by unix and wasi now, the target-first early return is still there, and the tests run on wasi too. One nit left: the comment in fsext/mod.rs:142 can just be // ctime isn't exposed by std Metadata outside unix. Also, does any CI job actually run the uucore tests on wasi? If not, the new wasi cfgs are still untested.

[eduardomourar]

Done. Added a step to wasi.yml in order to run the unit tests for uucore.